Research On Dynamic Reliability Quantitative Analysis Method Of DCS Based On BDMP

Instrumentation and control system(I&C)is the nerve center of nuclear power plant(NPP).Digital I&C system has been widely used inNPP,and its reliability evaluation is very important.The static event tree/fault tree method is the main method for reliability analysis of nuclear power plant system,which is not suitable for describing repairable system with component failure correlation.Markov model is more suitable to describe this kind of system,but it’s complexity increases exponentially with the increase of system components’ number.In practice,there are some limitations such as large model and poor readability.Boolean logic driven Markov processes(BDMP)is a high-level modeling method which can automatically generate Markov models.It can describe the repair behavior of device and the correlation between component failures.It has achieved a good balance between the model description ability and modeling simplicity.The BDMP analysis software YAMS can only use Monte Carlo simulation method for calculation.However,the accuracy of YAMS is lack of comparison and verification.In order to verify the effectiveness of YAMS,improve the efficiency of BDMP quantitative analysis algorithm,so as to deal with a larger scale BDMP model.A method for model transformation from BDMP to Markov Chain model is proposed,and the transformed Markov model is quantitatively analyzed by using probability model checker and the corresponding transformation and quantitative analysis software is developed.Further more,Erlang distributions is introduced to leaf nodes of BDMP to approximate period testing and repair.The developed software is applied to the water level control system of the steam generator.The results show that the proposed method can be applied to the medium-sized system,and the analysis result derived by the extended BDMP is more reasonable.